Developmental pattern of plasminogen activator activity in chick optic lobe

Neurochemical Research, Vol. 23, No. 9, 1998, pp. 1185-1190

Developmental Pattern of Plasminogen Activator Activity in Chick Brain Hemispheres G. Scicolone,1 S. Pereyra-Alfonso,1 J. L. Ferran,1 and V. Flores1'2'3 (Accepted January 15, 1998)

Plasminogen activators play key roles in several developmental events. In previous works we demonstrated the existence of typical developmental patterns of protease activity in the chick optic lobe and cerebellum. The aim of this work is to study the temporal pattern of development of plasminogen activator activity in the brain hemispheres. Plasminogen activator activity was assayed in soluble fractions derived by ultracentrifugation from Triton X-100 treated membrane fractions by using a radial fibrinolytic assay. Employing different inhibitors and anti-plasminogen activators antibodies we showed that developing brain hemispheres express only one type of enzyme which corresponds to the urokinase-type. Other results indicate that the protease activity displays a temporal pattern which completely differs from those of general parameters of development. This suggests that the plasminogen activator activity is developmentally regulated and could display specific functions during particular stages of development. KEY WORDS: Plasminogen activator; urokinase-type plasminogen activator; brain hemispheres; chick embryo; development.

INTRODUCTION

ture, as u-PA is involved in the control of cell migration, movement of cell processes and tissue remodelling (3). This proteolytic system is delicately regulated by a variety of mechanisms involving specific plasmin inhibitors (A2 antiplasmin) (2) and PA inhibitors (PAIs) (3), unspecific PA inhibitors -nexins- (4), proenzymatic forms of PAs (5,6) and membrane receptors that focalize the proteolytic action (7). Fairly recently, PAs have been implicated in a variety of biological events related to the development and overall maintenance of homeostasis within the central nervous system (CNS) (8). These molecules are considered to play important roles in processes controlling neu-

Plasminogen activators (PAs) are serine proteases synthesized and secreted by many different cell types. PAs activate a cascade of proteolytic activities that allows the PA-producing cells to degrade specific components of the extracellular matrix (1,2). Two different molecular types of PAs, tissue-type PA (t-PA) and urokinase-type PA (U-PA) were identified in several species. t-PA is primarily implicated in the overall maintenance of hemostasia within the vascula1

Institute of Cell Biology and Neurosciences "Prof. Eduardo De Robertis", School of Medicine, University of Buenos Aires, Paraguay 2155, 2nd Floor. 1121-Buenos Aires, Argentine. 2 University Institute of Medical Sciences, Favaloro Foundation. Solis 453, 1078-Buenos Aires, Argentine. 3 Address reprint requests to: Vladimir Flores; Instituto de Biologia Celular y Neurociencias "Prof. Eduardo De Robertis", Facultad de Medicina, Universidad de Buenos Aires. Paraguay 2155, 2° Piso, 1121-Buenos Aires, Argentina. Telephone number: 01-961-5010 int. 21. e-mail: [email protected].

ABBREVIATIONS: BHs, Brain hemispheres; CMF, Crude membrane fraction; CNS, Central nervous system; Extl, Extraction 1; E, Embryonic days; HH, Stages of development by Hamburger and Hamilton; NH, Newly hatched; P, Post-hatching days; PA, Plasminogen activator; PAI, Plasminogen activator inhibitor; SF, Soluble fraction; TH, Total homogenate; t-PA; Tissue-type plasminogen activator; u-PA, Urokinase-type plasminogen activator.

1185 0364-3190/98/0900-1185S15.00/0 © 1998 Plenum Publishing Corporation

1186 ronal and glial cell migration during development (9,10) as well as neuritogenesis (10,11). In previous papers (12-15) we demonstrated the existence of high levels of PA activity in the developing chick optic lobe and cerebellum. We demonstrated that, in these CNS areas, the PA activity displays a typical pattern of development which temporally correlates with periods of massive neuronal migration and also with lapses when neurite outgrowth and synapse formation and maturation are dominant events (14,15). This suggests that, apart from being involved in cell migration and neurite outgrowth, PA could also play a role in other developmental events such as synaptogenesis, synaptic maturation and plasticity (14,15). The present work aims at detecting the presence of PA activity in the developing chick telencephalon. Our purpose was to immunologically characterize the enzyme expressed in this major CNS organ and to determine whether the enzyme activity displays a typical developmental pattern during the embryonic and postnatal development. We have extended this study on PA activity to embryonic and postnatal stages of the developing brain hemispheres in order to clearly separate peaks of PA activity associated with processes of neuronal migration from peaks of PA activity associated with processes involved in neuronal circuits remodelling or refinement.

EXPERIMENTAL PROCEDURE Chemicals and Reagents. Agar was purchased from Difco, Detroit, MI. Human fibrinogen (Lot E 356) was a gift from Diagnostic Transfusion (Paris, France). Thrombin and aprotinin were gifts from Immuno, Buenos Aires. Trisma Base, amiloride, human plasminogenfree fibrinogen and Triton X-100 were purchased from Sigma, St. Louis, MO. Human urokinase was purchased from Serono, Buenos Aires. Reference t-PA from Bowes melanoma was a gift from Dr. D. Collen (University of Leuven, Belgium). Rabbit anti-mouse urokinase antibody (16) and goat anti-human Bowes melanoma t-PA IgG were gifts from Dr. L. Ossowski (Rockefeller University, NY, USA). Monoclonal antibody to avian u-PA, DMC-8, was a gift from Dr. J. Quigley (State University of New York, Stony Brook, USA) (17). Animals. White Leghorn chick embryos selected at defined stages of development, newly hatched, early postnatal and adult chicks were used. Eggs were incubated at 39°C and 60% relative humidity. From hatching onwards chicks were bred under normal cycles of 12 hr light and 12 hr darkness and supplied with food and water ad libitum. Brain hemispheres (BHs)—without optic and olfactory lobes and without cerebellum and brain stem—obtained from E10, E12, E14, E16 and E18 (E: embryonic days) embryos, newly hatched (NH), P3, P6, P9 and P12 (P: post-hatching days) and adult (45 days) chicks were studied. Developmental stages were recorded according to Hamburger and Hamilton's table of stages (HH) (18). BHs were removed, dissected free of the overlying membranes and weighted before processing.

Scicolone, Pereyra-Alfonso, Ferran, and Flores PA Extraction Procedure. Two to eight BHs obtained at characteristic stages of development were pooled for each assay. Total homogenates (THs) were prepared in 4 volumes of 0.1 M Tris-HCl buffer, pH 8.1 and immediately afterward ultracentrifuged at 100,000 g for 1 hr to separate a soluble fraction (SF) and a crude membrane fraction (CMF). The CMF was resuspended in the same buffer containing 0.5% Triton X-100 and immediately afterward ultracentrifuged at 100,000 g for 1 hr to obtain a soluble fraction containing the PA activity solubilized from the CMF with the Triton X-100 treatment (first extraction: Extl). All the procedures before described were carried out at 04°C. The Extl fraction was stored at -20°C. According to previous studies (19), this fraction allows to obtain reliable estimations of the total and specific enzyme activities. The Extl fraction was used for determination of PA activity as well as for biochemical and immunological characterization of the enzyme activity. For a detailed description see reference 19. Fibrinolytic Assay. After thawing, samples of Extl fractions were incubated for 1 hr at room temperature and then assayed for protein content and PA activity. Protein content was determined by the Lowry method (20) and PA activity was measured by a radial fibrinolytic assay in fibrin-agar plates (21) containing 5 mg/ml plasminogen-rich fibrinogen, 2.5% agar and 1 IU/ml of thrombin. The same batch of plasminogen-rich fibrinogen was used for all determinations. For a more detailed protocol description see reference 22. Experimental Controls. Plasminogen-free fibrin-agar plates were used to detect nonspecific protease activity in samples of Extl fractions obtained from E12, E18 and P3 chicks BHs. To investigate the presence of acid-labile PA inhibitors, samples of Extl fractions obtained at the above indicated stages were acidified to pH 2.5 with 1 M HO, incubated for 4 hr at 4°C and then neutralized to their original pH with 1 M NaOH (23). These samples were then assayed for PA activity. Characterization of PA Activity. In order to biochemically characterize the PA activity found in the developing BHs, samples of Extl fractions obtained from El2, El8, and P3 chicks were treated with different reagents which are known to possess specific effects on PA activity. The following studies were performed: /. Inhibition of PA Activity by Aprotinin. Samples of Extl fractions obtained from El2, El8, and P3 BHs were mixed with equal volumes of the same buffer solution containing appropriately diluted aprotinin (100 (MM) and incubated for 2 hr at 37°C (24). These samples were then assayed for enzyme activity. 2. Inhibition of PA Activity by Amiloride. Samples of Extl obtained from El2, El8, and P3 BHs were mixed with equal volumes of the same buffer solution containing appropriately diluted amiloride (100 uM) and incubated for 45 min at 37°C (25). These samples were then assayed for enzyme activity. 3. Inhibition of PA Activity by Non-Species-Specific Antibodies. Samples of Extl obtained from E12, E18, and P3 BHs were mixed with equal volumes of the same buffer solution containing increasing concentrations of anti-human t-PA and anti-mouse u-PA antibodies (16) and incubated for 1 hr at 4°C. These samples were then assayed for enzyme activity. 4. Inhibition of PA Activity by a Species-Specific Monoclonal Antibody. Samples of Extl obtained from El2, El8, and P3 BHs were mixed with equal volumes of the same buffer solution containing increasing concentrations of DMC-8, a monoclonal antibody that specifically inhibits the catalytic domain of chicken u-PA (17), and incubated for 1 hr at 4°C. These samples were then assayed for enzyme activity.

Plasminogen Activator Activity in the Developing Chick Brain

Fig. 1. Enzyme activity (area of lysis: mm2) in the Ext 1 fraction from El8 chick BHs as a function of time of incubation. Each point represents the mean +/- SD of three independent determinations performed in triplicate. In all figures SD is indicated only where it exceeds the size of the symbol.

Fig. 2. PA enzyme activity (mlU/ml) measured in the Ext 1 fraction obtained from El8 chick BHs along increasing protein concentrations. Each value represents the mean +/- SD of three independent determinations performed in triplicate.

RESULTS 1. Experimental Controls Enzyme Activity as a Function of Time of Incubation and Protein Concentration. Fig. 1 and 2 show linear correlations between protease activity detected in Extl samples of El8 and both, time of incubation and protein concentration respectively. Similar results were obtained by using Extl samples obtained from E12 and P3 BHs

1187

Fig. 3. Effect of plasminogen depletion, acid treatment and addition of aprotinin or amiloride on the enzyme activity found in Extl fraction obtained from El8 chick BHs. Each value represents the mean ± SD of three independent determinations performed in triplicate. Pig: plasminogen, treat.: treatment.

(not shown). Data obtained from these preliminary experiments were used to define fixed assay conditions. Time of incubation was fixed at 24 ± 2 hs and protein concentration was fixed at 8 ± 2 mg/ml for all the following experiments described in this work. Unspecific Fibrinolytic Activity and Effect of Acid Treatment. Fig. 3 shows that the enzyme activity found in Extl fractions obtained from El8 chick BHs completely disappears when samples are seeded on plasminogen-free fibrin-agar plates. Similar results were obtained using Extl fractions from E12 and P3 chick BHs. Given that no unspecific protease activity is expressed in the absence of plasminogen, the enzyme activity found in Extl fractions is, in fact, plasminogen-dependent. Fig. 3 also shows that the enzyme activity present in Extl samples from El8 chick BHs remains unchanged after the acid treatment. Extl fractions obtained from E12 and P3 chick BHs gave similar results. This indicates the absence of acid-labile inhibitors in Extl samples from the three stages of development studied. 2. Characterization of Enzyme Activity by PA Inhibitors Fig. 3 shows that the addition of aprotinin or amiloride, which are considered to be selective inhibitors of u-PA (24,25), to Extl fractions obtained from E18 chick BHs completely abolishes the enzyme activity. The same results were obtained using Extl fractions from E12 and P3 chick BHs.

1188

Fig. 4. Immunological characterization of enzyme activity. Inhibition of the enzyme activity found in Extl fractions from El8 chick BHs by non-species-specific antibodies (anti-human t-PA and anti-mouse u-PA) and by the specific anti-chick u-PA monoclonal antibody (DMC-8).

Scicolone, Pereyra-Alfonso, Ferran, and Flores were obtained with Extl samples from El2, El8, and P3 chick BHs. 4. Temporal Pattern of General Parameters of Development and of PA Activity Control experiments were performed to look for possible correlations between developmental pattern of PA activity and some general parameters of development such as BHs wet weight and protein concentration. BHs wet weight and BHs protein concentration, measured in THs and Extl fractions, show a regularly increasing profile along the embryonic and early postnatal development (not shown). The temporal evolution of PA activity depicted in Fig. 5 shows a completely different pattern. It displays an early embryonic phase of slow increase, between E10 (HH36) and E14 (HH40), followed by a later embryonic phase characterized by a sharp increase until hatching. Postnatally the enzyme activity abruptly decrease until P3. This is followed by a transient period of increased PA activity between P3 and P9 with maximal values at P6 and P9. From P9 onwards the activity declines to levels below the limit of resolution in adult BHs.

DISCUSSION

Fig. 5. Graphic representation of PA activity measured in Extl fractions as a function of time of development. Each point represents the mean ± SD of three independent determinations performed in triplicate.

3. Immunological Characterization of the Enzyme Activity Fig. 4 compares the effects of three different antiPA antibodies on the enzyme activity found in Extl fractions. An antibody directed to human t-PA has no detectable inhibitory effect whereas an antibody directed to mouse u-PA displays only a partial inhibitory activity. By contrast, a monoclonal antibody directed to chick uPA (DMC-8), which is considered to be an specific inhibitor of the catalytic site of the enzyme (17), completely abolishes the enzyme activity. Similar results

In previous papers we demonstrated the existence of typical developmental patterns of PA activity in the chick embryo optic lobe (12,14) and in the chick embryo cerebellum (15). This activity corresponds to u-PA and temporally correlates not only with lapses of massive neuronal migration but also with periods when neurite outgrowth and synapse formation and plasticity are prominent events (14,15,19). This suggests that, apart from being involved in cell migration and neurite outgrowth, PA could also play a role in other developmental events such as synaptic formation, maturation and plasticity. In order to immunologically characterize the PA activity expressed in the developing brain hemispheres and to investigate whether this activity displays a typical developmental pattern, we used the Extl fraction since, according to previous studies (19), this fraction gives better estimations of the total and specific enzyme activities. Characterization of the Enzyme Activity. Experimental controls demonstrated the absence of unspecific fibrinolytic activity and acid-labile inhibitors in samples of Extl fractions. These data indicate that the activity measured by the radial fibrinolytic assay is, in fact, the result of a plasmin-mediated activity on the fibrin-agar plate.

Plasminogen Activator Activity in the Developing Chick Brain The fact that amiloride, which is considered to be an specific u-PA inhibitor (25), completely abolishes the enzyme activity found in the developing chick brain hemispheres strongly suggests that the activity corresponds to a u-PA-like enzyme. Experiments of immunological neutralization of the activity by means of different specific antibodies corroborate this assumption. In fact, a complete inhibition takes place only with a monoclonal antibody that specifically inhibits the catalytic site of the chick u-PA (17). Our experiments of biochemical and immunological characterization of the enzyme, performed at three different developmental stages (El2, El8, and P3), give the same set of results. On this bases it can be postulated that only one type of PA molecule is expressed in the developing chick brain hemispheres. These results are consistent with previous reports indicating that the chick optic lobe—a mesencephalic derivative—and the chick cerebellum—a rombencephalic derivative—only express u-PA during the embryonic development (14,15,19). Interestingly, the chick neural crest cells, which are precursors of peripheral nervous system neurons, also express u-PA during the ontogeny (26). Correlations Between PA Activity and Developmental Events. The developmental profile of PA activity significantly differs from those of the general parameters of development. These results indicate that the developmental pattern of PA activity can not be accounted for by a gradual accumulation of an enzyme accompanying an undiscriminated increase in protein synthesis. By contrast, the typical fluctuations in PA activity that take place during the embryonic and early postnatal development followed by a gradual decrease until the adult stage suggest a developmentally regulated mechanism of PA expression. On these bases it can be postulated that PA activity could subserve specific ructions during particular stages of development. The developmental profiles of PA activity in chick optic lobe and cerebellum are completely different from that found in the brain hemispheres. Both of them display two peaks of high PA activity during the embryonic life. The first one occurs between E10 and E12 and the second one between E16 and E18 (14,15). The similarity between PA activity profiles in optic lobe and cerebellum is rather surprising at first sight. There is however acceptable explanation for this similarity by establishing temporal correlations between peaks of PA activity and some particular developmental events that take place in both structures (See references in 14 and 15). The role of PA activity in the developing nervous system is mainly associated with postmitotic neuronal migration (9,10) and, in fact, the first peak of PA activity

1189

in both optic lobe and cerebellum temporally coincides with a period of massive neuronal migration (14,15,2730). With regards to the second peak of PA activity, it must be mentioned that in the developing cerebellum, it corresponds to a period characterized by intensive circuits remodelling but this process overlaps with a second peak of massive neuronal migration (15,29-31). The second peak of PA activity in the developing optic lobe corresponds to a period clearly dominated by events involved in neurite outgrowth and synaptic formation and/or maturation, but there is still some processes of neuronal re-arrangement in the outermost neuronal layers of the tecrum opticum (14,27,28,32). We have extended our study on PA activity to the embryonic and postnatal stages of the developing brain hemispheres in order to clearly separate peaks of PA activity associated with processes of neuronal migration from peaks of PA activity associated with processes involved in neuronal circuits remodelling or refinement. Our experiments show the occurance of a first peak of PA activity during the ontogenetic life followed by an abrupt decrease between hatching and P3 and a second, smaller peak of PA activity, that takes place postnatally between P6 and P9. There is no massive cellular migration during this postnatal period. By contrast, it is known that intensive learning of environmental features takes place during the early postnatal life. In fact, there is substantial amount of experimental results documenting the ocurrence of a number of structural and neurochemical changes in the forebrain synapses and connectivities during the first 1 to 2 weeks after hatching (33-36). This period is clearly characterized by neural plasticity in the absence of any massive neuronal migration. In our opinion, in the developing forebrain, it can clearly be identified a peak of PA activity that exclusively correlates with a process of neuronal circuits remodelling and not with neuronal migration. This interpretation coincides with other observations which suggest the participation of proteases in synaptic remodelling (37-39). On these bases, the postnatal developmental pattern of PA activity could be a reliable parameter to study neural plastic changes induced by modifications in the environmental stimulation.

ACKNOWLEDGMENTS This work was supported by grants from the Consejo Nacional de Investigaciones Cientificas y Tecnicas (CONICET) and from the Secretaria de Ciencia y Tecnica, Universidad de Buenos Aires (Repiiblica Argentina).

1190 REFERENCES 1. Dano, K., Andreasen, P. A, Grondahl-Hansen, J., Kristensen, P., Nielsen, L. S., and Skriver, L. 1985. Plasminogen activators, tissue degradation and cancer. Advnc. Cancer Res. 44:140—266. 2. Hekman, C. M., and Loskutoff, D. J. 1987. Fibrinolytic pathways and the endothelium. Semin. Thromb. Hemost 13:514-527. 3. Hart, D., and Rehemtulla, A. 1988. Plasminogen activators and their inhibitors: Regulators of extracellular proteolysis and cell function. Biochem. Physiol. 90B:691-708. 4. Baker, J. B., and Gronke, R. S. 1986. Protease nexins and cellular regulation. Semin. Thromb. Hemost. 12:216-220. 5. Petersen, L. C., Lund, L. R., Nielsen, L. S., Dano, K., and Skriver, L. 1988. One-chain urokinase-type plasminogen activator from human sarcoma cells is a proenzyme with little or no intrinsic activity. J. Biol. Chem. 263:11189-11195. 6. Ranby, M., Bergsdorf, N., and Nilsson, T. 1982. Enzymatic properties of the one- and two-chain form of tissue plasminogen activator. Thromb. Res. 27:175-183. 7. Stetler-Stevenson, W. G., Aznavoorian, S., and Liotta, L. A. 1993. Tumor cell interactions with the extracellular matrix during invasion and metastasis. Annu. Rev. Cell. Biol. 9:541-574. 8. Murphy, P. G., and Hart, D. A. 1991. Regulation of plasminogen activator and plasminogen activator inhibitor expression by cells of neural origin. Semin. Thromb. Hemost. 17:268-275. 9. Moonen, G., Grau-Wagemans, M. P., and Selak, I. 1982. Plasminogen activator-plasmin system and neuronal migration. Nature. 298:753-755. 10. Shea, T. B., and Beermann, M. L. 1992. Regulation of neuronal migration and neuritogenesis by distinct surface proteases. Relative contribution of plasmin and a thrombin-like proteases. FEBS Lett. 307:190-194. 11. Pittman, R. N., Ivins, J. K., and Buettner, H. M. 1989. Neuronal plasminogen activators: cell surface binding sites and involvement in neurite outgrowth. J. Neurosci. 9:4269-4280. 12. Flores, V., Pereyra-Alfonso, S., Bal de Kier Jofre, E., and Fiszer de Plazas, S. 1988. Plasminogen activator activity in the developing chick optic lobe. Med. Sci. Res. 16:927-928. 13. Pereyra-Alfonso, S., Scicolone, G., Sato, M., Shu, H.-W., and Flores, V. 1991. Zymographic analysis of the plasminogen activator enzyme in the developing chick optic lobe. Com. Biol. (BA). 4: 299-304. 14. Pereyra-Alfonso, S., Scicolone, G., Ferran, J. L, Pecci Saavedra, J., and Flores, V. 1997. Developmental pattern of plasminogen activator activity in chick optic lobe. Int J. Devi. Neuroci. 15:805812. 15. Scicolone, G., Pereyra-Alfonso, S., Sanchez, V., and Flores, V. 1997. Temporal pattern of plasminogen activator activity in the developing chick cerebellum. Int. J. Devl. Neurosci. (In press). 16. Marotti, K, Belin, D., and Strickland, S. 1982. The production of distinct forms of plasminogen activator by mouse embryonic cells. Dev. Biol. 90:154-159. 17. Sullivan, M. L., and Quigley, J. P. 1986. An anticatalytic monoclonal antibody to avian plasminogen activator. Its effect on behavior of RSV-transformed chick fibroblasts. Cell 45:905-915. 18. Hamburger, V., and Hamilton, H. L. 1951. A series of normal stages in the development of the chick embryo. J. Morphol. 88: 49-92. 19. Pereyra-Alfonso, S., Scicolone, G., Fiszer de Plazas, S., Pecci Saavedra, J., and Flores, V. 1995. Current Triton X-100 treatments do not allow a complete plasminogen activator extraction from developing nervous tissue. Neurochem. Res. 20:137-142. 20. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, P. J. 1951. Protein measurement with Folin phenol reagent. J. biol. Chem. 193:265-275.

Scicolone, Pereyra-Alfonso, Ferran, and Flores 21. Angles-Cano, E., Boyer, B., Gisselbrecht, S., and Debre, P. 1984. Heterogeneity of plasminogen activator expression in various Moloney-virus-induced tumor cell lines. Lack of correlation with tumor growth and cell phenotype. Int. J. Cancer. 33:277-280. 22. Pereyra-Alfonso, S., Haedo, A., and Bal de Kier Joffe. 1988. Correlation between urokinase-type plasminogen activator production and the metastasizing ability of two murine mammary adenocarcinomas. Int. J. Cancer. 42:59-63. 23. Tissot, J.-D., Hauert, J., and Batchmann, F. 1984. Characterization of plasminogen activators from normal human breast and colon and from breast and colon carcinomas. Int. J. Cancer. 34:295-302. 24. Lottenberg, R., Sjak-Shie, N., Fazleabas, A. T., and Roberts, R. M. 1988. Aprotinin inhibits urokinase but not tissue-type plasminogen activator. Thromb. Res. 49:549-556. 25. Vassali, J.-D., and Belin, D. 1987. Amiloride selectively inhibits the urokinase-type plasminogen activator. FEBS Lett. 214:187191. 26. Valinsky, J. E., and Le Douarin, N. M. 1985. Production of plasminogen activator by migrating cephalic neural crest cells. EMBO. J. 4:1403-1406. 27. LaVail, J. H., and Cowan, W. M. 1971. The development of the chick optic tectum I. Normal morphology and cytoarchitectonic development. Brain Res. 28:391-419. 28. Scicolone, G., Pereyna-Alfonso, S., Brusco, A, Pecci Saavedra, J., and Flores, V. 1995. Development of the laminated pattern of the chick tectum opticum. Int. J. Devl. Neurosci. 13:845-858. 29. Hanaway, J. 1967. Formation and differentiation of the external granular layer of the chick cerebellum. J. Comp. Neurol. 131:114. 30. Otero, R. A., Sotelo, C., and Alvarado-Mallart, R. M. 1993. Chick-quail chimeras with partial cerebellar grafts: an analysis of the origin and migration of cerebellar cells. J. Comp. Neurol. 33: 579-615. 31. Ryder, E. F., and Cepko, C. L. 1994. Migration patterns of clonally related granule cells and their progenitors in the developing chick cerebellum. Neuron 12:1011-1029. 32. Williams, C. V., and McLoon, S. C. 1991. Elimination of the transient ipsilateral retinotectal projection is not solely achieved by cell death in the developing chick. J. Neurosci. 11:445-453. 33. Bradley, P. M., Horn, G., and Bateson, P. 1981. Imprinting: an electron microscopic study of the chick hyperstriatum ventrale. Exp. Brain Res. 41:115-120. 34. Horn, G., Bradley, P., and McCabe, B. J. 1985. Changes in the structure of synapses associated with learning. J. Neurosci. 5: 3161-3168. 35. McCabe, B. J., and Horn, G. 1988. Learning and memory: regional changes in N-methyl-D-aspartate receptors in the chick brain after imprinting. Proc. Nam. Acad. Sci. U.S.A. 85:28492853. 36. Horn, G. 1991. Imprinting and recognition memory: a review of neural mechanisms. Pages 219—261, in Andrew, R. J. (eds.), Neural and behavioral plasticity: the domestic chick as a model. Oxford University Press, Oxford. 37. Hantai, D., Rao, J. S., and Festoff, B. W. 1988. Serine proteases and serpins: their possible roles in the motor system. Rev. Neurol. (Paris). 144:680-687. 38. Qian, Z., Gilbert, M. E., Colicos, M. A., Kandel, E. R., and Kuhl, D. 1993. Tissue-plasminogen activator is induced as an immediately-early gene during seizure, kindling and long-term potentiation. Nature. 361:453-457. 39. Liu, Y., Fields, R. D., Festoff, B. W., and Nelson, P. G. 1994. Proteolytic action of thrombin is required for electrical activitydependent synapse reduction. Proc. Natn. Acad. Sci. USA. 91: 10300-10304.